[
https://issues.apache.org/jira/browse/HBASE-10051?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13834000#comment-13834000
]
Samir Ahmic commented on HBASE-10051:
-------------------------------------
Well, looks like there are some inconsistency in that comment and what is
written to zookeeper. Here is output from zkcli on my cluster:
[zk: node1:2181(CONNECTED) 0] ls /hbase
[meta-region-server, backup-masters, table, draining, region-in-transition,
table-lock, running, balancer, master, namespace, hbaseid, online-snapshot,
replication, splitWAL, recovering-regions, rs]
> rolling-restart.sh have old value of "zookeeper.znode.unassiged" config
> causing infinite loop
> ---------------------------------------------------------------------------------------------
>
> Key: HBASE-10051
> URL: https://issues.apache.org/jira/browse/HBASE-10051
> Project: HBase
> Issue Type: Bug
> Components: scripts
> Affects Versions: 0.96.0
> Environment: x86_64 GNU/Linux
> Reporter: Samir Ahmic
> Attachments: HBASE-10051.patch
>
>
> While running rolling-restart.sh without arguments or with --master-only
> argument script throws error when trying to check did all regions are cleared
> from transition state. Here is line causing error:
> {code}
> if [ "$zunassigned" == "null" ]; then zunassigned="unassigned"; fi
> {code}
> AFAIK default value of "zookeeper.znode.unassiged" is changed to
> "region-in-transition". So we should change this line to:
> {code}
> if [ "$zunassigned" == "null" ]; then zunassigned="region-in-transition"; fi
> {code}
> I will attach patch today.
--
This message was sent by Atlassian JIRA
(v6.1#6144)
